AI advancements in diabetes diagnosis promise better healthcare

Researchers at Stanford University are exploring the potential of artificial intelligence (AI) in the diagnosis of diabetes, leading to advancements that could provide better and more accessible healthcare for patients. This research delves into the complexities of Type 2 diabetes, which constitutes approximately 95% of diabetes diagnoses, unveiling important subtypes that may predict associated risks for conditions such as kidney, heart, or liver diseases.

Dr. Tracey McLaughlin, an endocrinology professor at Stanford, pointed out the limitations of traditional metabolic testing, stating, “Understanding the physiology behind [diabetes] requires metabolic tests done in a research setting, but the tests are cumbersome and expensive and not practical for use in the clinic.” In light of this, the research team employed data from glucose monitors to develop an innovative algorithm capable of identifying three of the four most prevalent subtypes of Type 2 diabetes.

The algorithm demonstrated a significant advantage over conventional clinical data, as it “predicted metabolic subtypes, such as insulin resistance and beta-cell deficiency, with greater accuracy than the traditional metabolic tests”—achieving this with an impressive 90% accuracy rate. Identifying a patient’s specific subtype of diabetes holds critical implications for tailoring treatments. Through this categorisation, healthcare providers can design personalised medicine plans, optimising resource allocation and potentially lowering expenses.

According to McLaughlin, “This matters, because depending on what type you have, some drugs will work better than others.” The research aims to facilitate a more convenient and immediate means for individuals to gain insights into their health, particularly those who might lack access to conventional healthcare resources due to geographic, economic, or other barriers.

With around 13% of the population in the United States diagnosed with diabetes, recognising these treatment nuances could greatly influence the effectiveness of various therapeutic approaches. The ability of AI to extract deeper insights from data sourced from devices that patients already utilise, such as glucose monitors, underscores the technology’s potential to enhance healthcare delivery.

This breakthrough comes on the heels of significant recognition at CES 2025, where two over-the-counter glucose monitors received accolades in the Digital Health category. As research continues to evolve, leading to greater capabilities in AI-driven healthcare, the hope is to increase the availability and precision of health information for individuals across diverse circumstances, contributing to overall better health outcomes.

Source: Noah Wire Services